This article examines how poverty was defined as the core of the new social question in Argentina during the democratic transition. In order to do so, we analyse how the definition and classification of poverty carried out by experts at the National Institute of Statistics and Censuses (Instituto Nacional de Estadística y Censos, INDEC) in the 1980s contributed to this process. At the time, several Latin American countries embarked on constructing poverty as an object of statistical measurement and of state action. While the causes and effects of ‘marginality’, a conceptual innovation that arose from the Latin American debates over poverty, were still under discussion, the launch of some new technical initiatives in the late 1970s and early 1980s appeared to reflect the emergence of an unprecedented interest in quantifying poverty, which foreshadowed the region's turn towards developing its own tools to describe and interpret this phenomenon.Footnote 1 , Footnote 2
The Argentine case is relevant for various reasons. First of all, it constitutes a ground-breaking effort within the Latin American region: for the very first time, a state statistical agency attempted systematically to test a method of measuring poverty by considering basic needs, while promoting the implementation of similar measures in other countries within the region as well.Footnote 3 The ties between experts from these countries as well as the impetus of international institutions such as the Economic Commission for Latin America and the Caribbean (ECLAC) helped to expand initiatives focusing on poverty measurement. Although in the 1990s measurement by income progressively displaced measurement by basic needs, INDEC's statistical innovation contributed to establishing new ways of reading the social question, as part of a new repertoire of social knowledge that was adopted by Latin American states. Second, the change of political regime also contributed to the emergence of these statistical tools for two reasons: (a) because of the political interest of the new government in quantifying the ‘damages’, to use the terminology of the time, caused by the dictatorship, and (b) by virtue of the fact that the need for social reparation to amend the authoritarian ‘heritage’ oriented the decisions of the experts (most of whom had recently returned from exile) who were responsible for producing the first measurements of poverty. Finally, the Argentine case allows us to grasp particularly clearly how a statistical object, in order to help classify and make intelligible the social world or some of its spheres, requires a convergence between the corroboration of the emergence of a new social phenomenon, the political and moral concerns that contribute to framing it as an object of public policy at a certain critical juncture in history, and the activity of a group of experts who are able to mobilise effort, knowledge, institutional and personal resources, as well as international networks in order to elaborate a new measurement tool.
In methodological terms, we studied this process through the reconstruction of the professional trajectories of experts at INDEC, the identification of the resources these agents mobilised, and the analysis of their intellectual production. Our research drew upon four types of sources: (i) interviews with the agents in question and other specialists in poverty and its measurement in Argentina and Latin America; (ii) local archives of national public offices (Sistema de Información, Monitoreo y Evaluación de Programas Sociales/SIEMPRO, INDEC, Ministry of Economy) and multilateral agencies (ECLAC, United Nations Development Programme/UNDP, Inter-American Development Bank/IDB, World Bank); (iii) documents of the research centres that hosted the experts-scholars involved in measuring poverty and designing and evaluating social policies, and (iv) press archives.Footnote 4 The analysis of the documents produced by the INDEC between 1980 and 1990 on poverty, as well as of those published by ECLAC, UNDP, the IDB, and the World Bank, allowed us to reconstruct the first measurements of poverty at the INDEC and the links it established with regional and global experiences and concerns regarding the matter. As a methodological strategy aimed at grasping the terms of public debate on poverty and its scope, we examined the two newspapers with the largest circulation in the country, Clarín and La Nación, as well as the important political weekly of the 1980s, the magazine Somos, by selecting an intentional sample that focused on the specific junctures at which the INDEC produced and made public its reports.
In the first place, the present article defines, in the light of the bibliographic debates of recent decades, the different approaches to poverty as an object of measurement and as a common language of public debate on social issues. Second, it reviews the ground-breaking experience of quantifying poverty at the time of reinstating democracy, paying particular attention to the formation of the expert group at the moment when poverty was publicly established as an object of state action and as a moral problem. It shows how the state demand for data that was necessary for the distribution of a food aid programme was transformed into an opportunity to create an instrument of public statistics that stabilised poverty as a measurable object. Afterwards, the article describes the second programme of poverty measurement developed at the INDEC, the Research on Poverty in Argentina (Investigación sobre Pobreza en Argentina, IPA), analysing its means of social validation and its contributions to the statistical accuracy and complexity of the phenomenon at a time when the deterioration of the living conditions of a certain sector of the country's population worsened, which led to the measurement of poverty becoming a subject of public interest. Finally, this article provides general conclusions about the legacy of these pioneer initiatives of measuring poverty in Argentina and in the region, and considers public controversies related to this topic.
Poverty as a Statistical Object
Peter Townsend has stressed the importance of considering the events and the historical circumstances that influenced the emergence of the measurement of poverty in a given country.Footnote 5 Given that the first attempts to quantify the poor date back to the end of the nineteenth century, it is pertinent to ask why this interest arose with such force in the Argentina of the 1980s.Footnote 6 Specifically, what factors contributed to the emergence of the first measurements of poverty in Argentine state agencies? How did the measurement of this phenomenon become socially accepted as one of the dominant modes of apprehending the social problems of the country?Footnote 7 It is thus a matter of analysing the actors, institutions, and knowledge that participated in the process of constructing the statistical instrument in question, as well as assessing its political impact.
Unlike perspectives that concentrate on the emergence of measured phenomena or on methodological debates concerning the accurate ways of measurement, we focus our reflection on the social process of producing statistics on poverty, taking them as cognitive tools for ordering the social world and its problems. In this sense, we follow in the line of research dedicated to the analysis of the performativity of expert activities that show that these activities, by defining objects of intervention and measurement, redefine social groups and social problems.Footnote 8 According to this approach, statistics are a means of reading social reality that contributes to configuring that same reality inasmuch as they produce the conventional objects that circulate as part of representations shared by society and thus nourish public debate.Footnote 9 Additionally, as several researchers working on the social construction of social problems argue, the forms of representation of reality and of social groups constructed by statistics promote a definition of public problems that dominate the political agenda.Footnote 10 In other words, statistics are involved in the framing of the problem as well as in the redefinition of the manner in which it is socially perceived, conceptualised and treated.Footnote 11 Statistics are also practical tools in the sense that they establish certain principles to orient public policy, and once the state consecrates their importance and instates them as principles of vision and division of the social world, they become powerful weapons of technical and political intervention in society.Footnote 12
Several recent articles underscore the importance of experts in defining public policies and social problems.Footnote 13 However, even though the literature has widely discussed the case of economists, showing that their ‘rise’ coincides with the emergence of a ‘lingua franca’ of finance and the global economy, specialists in statistics have not received much attention, and there has been no analysis so far of the implications of the fact that quantification is a resource used preferably by expert groups to reach and defend privileged positions in the democratic debate.Footnote 14 The process of building an expert domain on poverty measurement also produced a specific technical language, a characteristic typical of an internationalised technocratic elite.
Indeed, the institutionalisation of the measurement of poverty in Argentina would not have been possible without the social, technical and conceptual networks that supported this process. Our approach integrates the observation of a historical process of building technical capacities and accumulating technical expertise in a state agency with the study of instruments for reading the social world and of the ‘grammar’ generated by the language of statistics to define the social question in terms of poverty. Although techniques have changed significantly since then, the success of this enterprise of measurement can be seen in the fact that, in the current public debate, poverty appears as a ‘fact’, something taken for granted, quantified and standardised, even though the actual number of poor has become an object of controversy. Since the determination of those figures is used for evaluating government performance in the social sphere, statistics on poverty have become a subject of intense public debate concerning how accurately those measurements reflect the reality of Argentine society as well as regarding the quality of the data produced by the office of public statistics. In the past two decades, several ministers of economy, whose department the INDEC depends on administratively, made public their mistrust of the official numbers on poverty and attempted to control the production of statistical data on Argentine economy and society. Recently, the generalised lack of confidence in public statistics, a direct consequence of the political intervention in the INDEC by Néstor Kirchner's government in 2007, severely affected the credibility of statistics on poverty.Footnote 15 The political opposition attributed the absence of reliable statistics on poverty to a strategy promoted by the Argentine state (which refused to recognise the problem) to force the ‘social disappearance’ of the poor. This complaint expresses both the centrality of the statistical instrument in the construction of social problems as well as the performative capacity of numbers to grant (or deny) existence to a phenomenon based on its visibilisation in the public sphere.
Poverty and the Democratic Reparation
At the beginning of the democratic transition, along with issues that already formed part of the political agenda such as those linked to the labour market and housing, a new social problem began to be outlined in the public debate that made it possible to collectively refer to a number of different but related situations such as the deterioration of conditions of life, loss of income, and so on. At the INDEC, two projects helped to define the problem technically in terms of poverty. On the one hand, at the request of policy-makers engaged in designing a social assistance programme, the experts at the INDEC conducted a study based on data from the 1980 Census. Published as La pobreza en Argentina (Poverty in Argentina), it helped map the areas with the greatest concentration of poor households in the country. On the other hand, in 1987 the experts at the INDEC launched the IPA project, which had as one of its main objectives precisely to improve the quality of the measurement of poverty. Both initiatives point to how poverty became a central issue in the agenda of the official agency of statistics, and how it turned into a concern for the economists, sociologists and demographers who by then had rejoined the institute.
The new government and its policy-makers shared this concern regarding the deterioration of the living conditions of the masses. Indeed, president-elect Raúl Alfonsín had made poverty one of the axes of his campaign speeches.Footnote 16 Once he assumed office as president, one of his first actions was to implement a food aid programme of unprecedented breadth: the national food programme (Programa Alimentario Nacional, PAN; pan being the Spanish word for ‘bread’). Implemented by the Ministry of Health and Social Action, this programme, initially intended as transitional, became one of the main instruments of the government's social policy. The policy-makers commissioned experts at the INDEC to create a statistical rule that would geographically orient the distribution of the food aid.Footnote 17 This request provided an opportunity to develop the statistical category of ‘poverty’. Given the loosely coordinated nature of Argentine state agencies, the technicians at the INDEC enjoyed considerable autonomy. For them, it was:
[e]specially urgent to define, quantify, locate, and diagnose the different situations of poverty in which a significant part of Argentine households can be found. Thus the INDEC fulfils the fundamental and inevitable task of producing information that serves as a basis for the design and implementation of social policies by the Argentine Government.Footnote 18
Statistics helped delineate a new domain of state intervention while contributing to define a clientele for targeted social policies. Qualifying the quantification of the poor as a ‘fundamental’ and ‘inevitable’ task of a democratic government, the authors of the document laid down the moral position that they assumed, which was closely linked to their diagnosis (not universally accepted by society) that the problem was inherited from the previous authoritarian regime.Footnote 19
The objectives arising from the technical innovation of a new generation of professionals who had been trained to act as state agents were therefore combined with the political objective of social reparation of the new government, which understood that the democratic era had to make amends for the damage caused by the dictatorship, and that in order to do so the state had to equip itself with instruments capable of reading accurately social reality. Statistics therefore became an instrument of public action.
At the international level, multilateral agencies (especially ECLAC, and later the World Bank and UNDP) became interested in elaborating forms of poverty measurement at the regional level that would allow comparisons between the different realities of countries that, in some cases, had already adopted adjustment policies, and, consequently, promoted initiatives like the one designed at the INDEC. The goals of such initiatives included elaborating tools for diagnosing the social situation of the countries of the region within the context of an increasing transnationalisation of public policies.
Expert Trajectories: Crosses and Confluence
The working group that pioneered the measurement of poverty in Argentina was formed after the new authorities of the INDEC took office following the establishment of the Radical government in 1983. In 1984, Luis Beccaria, an economics graduate of the University of Buenos Aires (UBA) and a specialist in statistics, was appointed director of the agency, a position he held until 1990.Footnote 20 He was 34 years old at the time, and had a fluid relationship with Juan V. Sourrouille, Secretary of Planning at the Ministry of Economy at the beginning of Alfonsín's government, whose office the INDEC depended on administratively. They had known each other since the 1970s, and they had coincided recently at the Industrial Bank, a public bank where Beccaria was Head of Research.
In close collaboration with ECLAC, and pursuant to a demand from the new government to produce data that would identify the areas of the country with the highest levels of social deprivation, the new director of the INDEC proposed to begin a statistical study of the social situation ‘inherited’ from the dictatorship. In response to a request from the managers of the PAN, Luis Beccaria, together with Alberto Minujin, a mathematician with postgraduate studies in applied statistics and demographics, and Oscar Altimir, an economist at ECLAC, were the first to attempt to identify a phenomenon they defined as ‘poverty’.
The initial core group was thus led by experts who had already worked at the INDEC or at other state institutions in the 1960s and 1970s, moved by the moral and professional impetus that located the engine of the ideals of social change in state intervention, allowing them to combine their political commitment with a technical and professional career.Footnote 21 In fact, the joint work that started with the new democratic period was a new phase in their relationship: their trajectories had overlapped and their ties were forged during the earlier developmentalist period.
Altimir, like Sourrouille, had studied at the Faculty of Economic Sciences (Facultad de Ciencias Económicas, FCE) of the University of Buenos Aires, under the modernising influence of this academic space that aimed to provide the developmentalist state institutions with newly-trained professionals and experts. At the time, the social recognition of economists as the experts responsible for planning was on the rise, and even more so if their toolbox included the command of statistics and econometric models. The two experts coincided in different settings and in different functions. First, they attended Julio Olivera's courses on economic analysis and collaborated with him at the FCE.Footnote 22 Later, at the beginning of their careers, they worked for developmentalist organisations, such as the Federal Investment Council (Consejo Federal de Inversiones, CFI) and the National Development Council (Consejo Nacional de Desarrollo, CONADE), or in non-governmental institutions linked to the former organisations that were part of the process of the institutionalisation of economics as a professional field.Footnote 23
Another common feature of the trajectories of these professionals was their experience in multilateral agencies that participated in the consolidation of social science in the region. These agencies, ECLAC most notably, were spaces of circulation of knowledge, social scientists and state technical staff.Footnote 24 Sourrouille worked on a permanent contract at ECLAC in the early 1970s, whereas Altimir built there a lasting career as a researcher and eventually as director of statistics. It was in fact at ECLAC, in 1978, that Altimir led the first research project on poverty in Latin America, which aimed to quantify the phenomenon through surveys by sampling conducted in different countries of the region. There he defined a threshold of poverty for each country, trying to establish and synthesise minimum welfare standards that guaranteed a dignified subsistence according to the values projected by the style of development promoted by ECLAC.Footnote 25
Moreover, Beccaria and Minujin joined the statistics office after its creation as an institute (Law 17.622, 1968), during a phase of professionalisation promoted by Sourrouille while acting as director of the INDEC (which fell under the administrative control of the CONADE). Beccaria and Minujin joined the agency during a massive recruitment campaign of young people who had recently graduated in subjects then considered ‘modern’ (such as economics, sociology and statistics). This generated a new political climate at the INDEC, characterised by youth mobilisation, the practice of discussions in assemblies, as well as political activism. Minujin was employed in the area of demographic and social statistics. During the 1970s, he produced several documents on social statistics in Argentina at the INDEC in collaboration with ECLAC. While still an undergraduate student, Beccaria worked at the Department of Economic Statistics of the INDEC between 1973 and 1975. In the mid-1970s, he moved to England to study for his PhD in Economics at the University of Cambridge. The coup d’état in 1976 pushed Minujin into exile. He worked with Beccaria in England for a brief period. Afterwards, he continued his exile in Mexico similarly to many of his colleagues committed to the Peronist Left. In 1984 he returned to the INDEC as National Director of Social Statistics, and in 1987 he took charge of the direction of the IPA. Pablo Vinocur, who was then called upon to coordinate the project, shared with him the personal experience of exile as well as a characteristic feature of the group: they all capitalised on their years abroad to consolidate their careers. A sociologist trained at the UBA (he graduated in 1974), Vinocur also went into exile in 1976, but ended up in Costa Rica. His work in public health and social programmes enabled him to make contact with international organisations, in particular with ECLAC. When he returned to the country in 1983, he was able to reintegrate professionally into local offices of organisations such as the Organisation of American States and state agencies such as the Ministry of Social Action.Footnote 26
Thus, work in multilateral organisations and foreign universities emerges as a feature of the experience of exile that could be capitalised on in professional terms. Navigating these foreign spaces allowed the experts to return from their exile better prepared for the ups and downs of the world of experts contracted to work for the state and for moving around from the field of NGOs to international bureaucratic positions, from state agencies to research centres, with relative ease.
The idea of an international career should not be considered in these cases as a plan drawn up clearly in advance.Footnote 27 It was in part an unintended consequence of their exile that these specialists accumulated experiences and credentials that could be exploited later in other fields and national spaces. The proximity to the state was also part of a specific moral and professional commitment.
In 1984, the moral commitment to social reparation after the dictatorship was sufficiently broad and comprehensive to mobilise, under a Partido Radical government, developmentalist economists as well as social scientists with long-standing political activity within revolutionary Peronism. The conviction forged in their political and professional socialisation that linked social transformation to state action worked as a unifying factor. The experts trained in exile returned home in possession of the knowledge considered necessary to carry out this enterprise.
The Construction of a Statistical Instrument
With the democratic opening, ECLAC wanted to provide technical support to the new initiative at the INDEC. With this support, one of ECLAC's major experts in the matter, Oscar Altimir, managed together with Minujin and Horacio Somigliana the task of developing the new poverty statistics. As a way to take advantage of information from demographic and housing censuses, ECLAC promoted the application of one of the methods of measurement of poverty already tested in the world, the ‘unfulfilled basic needs’ method (NBI in Spanish) in the region.Footnote 28 This type of instrument is built by defining a series of indicators (overcrowding, precariousness of the habitat, lack of sanitary facilities, early desertion from the school system, etc.) usually reported by population censuses, allowing thus to determine whether specific households fail to meet any of the needs deemed essential to be identified as poor. Therefore, the instrument is based on determining past consumption, assuming an ‘ecological’ dimension of poverty.Footnote 29 This identification was at odds with the conceptual bases of the PAN, according to which acute food deficiency appeared as the core of the definition of poverty. The advantage was that, once the dissatisfaction of those needs was established, it could be used to elaborate ‘poverty maps’ and locate geographically the surveyed deficiencies with the exhaustiveness of the census information.
This technical decision was supposed to replace another approach, that of the poverty line (PL), which had had a significant presence since the 1970s, given the wide circulation of the revision by the English specialists Peter Townsend and Brian Smith of Charles Booth and Seebohm Rowntree's original proposal, and their criticism of the notion of subsistence.Footnote 30 Under the PL approach, poverty was associated with the impossibility of reaching a minimum threshold of consumption, a concept that is closer to the assumptions of the PAN.
However, the experts from INDEC defined poverty in terms of NBI.Footnote 31 The construction of the first device for the measurement of poverty in Argentina, rather than conforming to the categories that operated in the design of the programme that it was intended to serve, responded to political imperatives in its implementation: the census was the only source of comprehensive, reliable data that could be disaggregated across the entire national territory (a necessary condition for orienting the distribution of the PAN), but it did not contain questions about the income of households or individuals, which therefore excluded the possibility of a measurement based on such data. Adjustment of the possible indicators to the local reality was the focus of the internal discussions of the expert group and of the tests that were carried out during the process.Footnote 32 The indicators linked to the material conditions of housing (such as low-quality housing and overcrowding) had a singular relative weight among the indicators that were finally selected. In addition to reflecting technical factors, the sensitivity of the instrument by housing conditions needs to be examined from two angles: first, by considering the protagonism that the problems of housing had acquired in the political agenda and in public statistics,Footnote 33 and second, in view of the personal experience of territorial political militancy of certain technicians of the INDEC, who were worried about the existence of slums, marginal neighbourhoods, or enclaves of poverty, as they were conceptualised at the time.
The political imperative to assess the new social situation, on the one hand, and the lack of previous historical experiences that could have supported the discussion on poverty, on the other, gave a particular empirical bias to the process of elaborating the new index. According to Beccaria:
In Argentina the issue of poverty was never a central issue in the political agenda. There was concern about marginalisation and its associated concepts, but it was always more of a conceptual discussion, there had never been a major concern over the measurement of poverty. […] In this case we worked backwards. I have a concept, I try to operationalise it, and then I make an instrument to measure it. Here it was the other way around. I have the census, I have the instrument, […] and then I have to see how to make the most of it in order to approach the concept with the data that I have.Footnote 34
Although in 1979, in a pioneer publication by Altimir, the study of situations of deprivation still made use of other concepts more established in the academic discussions in the region, like those of marginality and informalityFootnote 35 (which attracted Beccaria as an object of purely academic interest), the institutionalisation of technical ways of measuring poverty ended up relegating those discussions to the background.Footnote 36 The debate on marginalisation had identified the situation of people in the labour market and their housing conditions as central dimensions of the problem. In parallel to the development of this debate, the INDEC had put considerable effort into classifying the various situations related to employment and into characterising the labour market in socio-demographic terms.Footnote 37 The construction of the first poverty measurement instrument in the 1980s, on the other hand, would be set in motion by another intuition: namely, that the dynamics of the labour market did not fully account for the workings of the social world, and that it was necessary to integrate other aspects concerning social relations that are articulated at the family and community level. The displacement of the conceptual discussion on marginality, although some traces persisted in the documents of the INDEC in the 1980s, meant getting closer to other types of positions that were present in the prevailing discussions in the field of economic science in the 1970s, in particular, the famous controversy between the economists Townsend and Amartya Sen on absolute and relative poverty.Footnote 38 Altimir was among those who introduced Sen's writings to Latin America at the end of that decade. They would go on to have a significant influence in the region in the early 1990s.Footnote 39 Nevertheless, the pioneer work carried out at the INDEC on poverty measurement shows traces of both conceptions.
Thus, as stated by the experts we interviewed, the construction of the concept of poverty was not one of the principal concerns of the technical mission.Footnote 40 Instead, the methodological debate prevailed. The imperatives of a recently sworn-in government that had to respond to political demands concerning a specific problem (namely, poverty) that had gained a place of its own in the political agenda forced the technicians to proceed ‘backwards’, as described above by Beccaria (that is, from data to the concept). The priority given to how to measure over what to measure could also be due to what experts took for granted by virtue of their background of professional experiences acquired abroad that they applied to Argentine reality, which had hitherto been scarcely investigated through this prism. Additionally, the impulse by ECLAC might have also contributed to the hasty advances in experimentation with certain instruments, this time at a state agency and facilitated by previous connections and shared interest among specialists. This transformation in the mode of conceiving, classifying, and therefore speaking about the social question seemed destined to be dominated by debates on how to measure poverty rather than defining it. The commotion caused by the publication of Poverty in Argentina, both in expert circles and in the public debate, shows the importance of the cognitive transformation that was taking place. The impact of the results of the measurement by NBI in the press reveals that Argentine society faced a novel situation, where ‘the deeper phenomenon of the last [i.e. previous] decade still escape[d] [social] consciousness’.Footnote 41 For a society that imagined itself as one of the most egalitarian of the region, poverty seemed a residual phenomenon. That is why its statistical incidence generated perplexity; compared to the numbers delivered by that first study by Altimir that had placed Argentina among the countries with the lowest rates of poverty in Latin America, the results of the INDEC report created a new public awareness of the phenomenon.Footnote 42 At the same time, the use of the category ‘poverty’ by the experts was an innovation. The usefulness of the concept still generated doubt among their peers, inclined to think of the problem in terms of already existing categories such as marginality or unemployment. Some specialists made public their objections against this first measurement. The economist Álvaro Orsatti, who would later participate in the IPA, pointed out that these figures constituted a ‘partial balance’ of the social situation and that the ‘economical-labour’ variable (absent in the new index) had an explanatory value of poverty at least as important or even more important than the NBI.Footnote 43
It was during the 1980s that the rough-and-ready notion of poverty (Altimir called it in 1979 ‘a grossly qualifying social category’) went on to refer to a well-delimited population.Footnote 44 Even though the already mentioned tensions and debates persisted, the associations between concepts, measuring instruments, actors and institutions became over time more or less stable, so as to establish the solidity of the numbers on poverty and consolidate it as a concept that gradually subsumed other notions (informality-marginality). Clearly, it was not only an intellectual matter, but it also constituted a sort of a recipe for action (a ‘black box’) that amalgamated ideas, actors and institutions, and at least in the Argentine case (in which social statistics were strongly associated to the labour world) represented a renewal of prevailing conceptions and actions about the social world, their problems and, with the object already well-established, its consequences for democratic regimes struggling with problems of consolidation.Footnote 45
The influence that the INDEC's experience had in the region also helped to solidify locally the statistical notion of poverty. The experience was exported to other Latin American countries through ECLAC, interested in disseminating the measurement of poverty not only as a way to produce reliable national social statistics, but also as a guarantee of technical uniformity, enabling the comparison and allowing the definition of common parameters to orient multilateral resources of ‘fighting against poverty’ to different countries. A project funded by the UNDP, co-managed by ECLAC and directed by Julio Boltvinik, was crucial for this purpose. It was called ‘Regional Project for Overcoming Poverty’, and from 1986 it organised a series of meetings of experts and officials from the field of statistics and social policy with the aim of sharing the measurement techniques used in each country. Altimir, Beccaria, Minujin, and other members of the Argentine team had a leading role in the meetings. According to Feres and Mancero, the INDEC's was one of the ‘decisive studies on method’ that exerted influence on the other participating countries that ‘resorted to the same indicators’.Footnote 46 Boltvinik, meanwhile, maintains that the measurement by NBI ‘became widespread in Latin America on the basis of the use the INDEC made of it’.Footnote 47
The Stabilisation of the Statistical Object in the Argentine Social Crisis
In 1986, through negotiations with the Ministry of the Economy, the team headed by Minujin managed to join a programme of state reform, funded by the World Bank, which implemented a project to improve the instruments for poverty measurement and to generate new lines of research on living conditions. In 1987, they created the IPA, recruiting a broad and multidisciplinary work-team, with specialists in several areas such as health, education, the young, among others. In 1988, Vinocur joined the direction of the programme, first under Minujin's command, later as head of the programme. Scholars and experts were especially hired for this research. This type of contractual relation with the state, based on professional fees paid for projects carried out within the framework of programmes financed by multilateral agencies, was to become the dominant form in the field of experts on poverty and social policies over the next decade. On the one hand, the external financing made it possible to rely on specialised staff and obtain the necessary equipment to carry out the project. On the other hand, this type of hiring allowed the payment of high fees, which were attractive for well-trained staff and for the international market of consulting, even though it did not foster the strengthening of state bureaucracies, and it resulted in very few technicians continuing to work within the state apparatus. It was an institutional modality that allowed experts to have flexible careers, though it transformed the state into a succession of layers of devices and people that were difficult to articulate.Footnote 48 The IPA was not an exception in this regard; it was relatively isolated from the INDEC, even in a physical sense, since it was located in another building.
Following the success of the first measurement, the objective of the new research was twofold: on the one hand, it tried to generate primary data by establishing a basic basket to measure poverty in relation to the income of households, thus combining the method of NBI with that of the PL. On the other hand, the experts sought to produce ‘information about the conditions and the characteristics that urban poverty assumes in Argentina, once the different existing situations were defined with respect to the structure of the satisfaction of the needs, the gravity of the deficiencies, and the perceptions of the affected groups concerning those deficiencies’.Footnote 49 The incorporation of a qualitative approach was a novelty for the INDEC; this may be associated with the turn towards qualitative research methods that took place in the Argentine social sciences in those years.Footnote 50
Here, the link with public action was also based on the legitimacy of the programme. The results of the IPA had to offer ‘quantitative elements to create or rebuild social policies’.Footnote 51 The document that presented the project clearly reflects a key factor in the stabilisation of poverty as a statistical object and as an instrument of public action: the phenomenon to be measured was no longer seen as a circumstantial inheritance of a past that should be left behind, but rather as a ‘reality’; ‘poverty became a reality of considerable importance for Argentine society as well as a presence that demands its inexorable eradication, especially since the victory of the democratic regime’.Footnote 52
Again, the combination between building expert careers and developing a moral commitment was central to the progress of the project. During the years of crisis of Alfonsín's government, both the Radical and Peronist experts (the latter already involved in a programmatic work in the technical team of the Peronist Party candidates for the 1989 presidential elections) remained committed to the project. In fact, the IPA was a space for producing professional loyalties; it created a certain pride in belonging that accompanied its members in later years. Most of them would meet again later in different ministries, state agencies, or international agencies, always drawing on that previous knowledge, valued positively, which made them recommend each other independently of their political allegiances.Footnote 53 Members of the IPA obtained significant professional benefit from this experience, which allowed them to continue their careers with a certain degree of distinction. Without a doubt, a great proportion of the staff that would later become experts on poverty and the measurement of poverty in Argentina was formed there, and they came to occupy the key relevant posts for the next decade.
The international links had already been consolidated. The relationship with ECLAC and with the UNDP's ‘Regional project for the overcoming of poverty’ transformed the IPA into the local version of a regional effort focused on measuring poverty, instead of it remaining an isolated programme. The project benefited from exchanges with experts from ECLAC of Chile, especially Altimir, Feres and Rolando Franco, with Boltvinik's team in Mexico, and Rubén Katzman's in Uruguay. They served as interlocutors for the IPA experts, with whom they were able to test their hypotheses and among whom they could disseminate their findings.
Less conditioned by the urgency of the distribution of social programmes, the experts who participated in the IPA began to work to measure poverty by income. They studied the structure of household budgets as well as the nutritional needs of adults and children in order to elaborate the basic food basket.Footnote 54 They defined then the minimum consumption of a family in order to set the income thresholds for indigence and poverty, to be corroborated against information from the Encuesta Permanente de Hogares (Permanent Household Survey, EPH); this survey would then become a source of information-gathering to determine the new indices of poverty.Footnote 55 , Footnote 56
One of the innovations of the programme was that it combined two types of poverty measurement: NBI and PL. In the IPA, experts worked on testing the different ‘sensitivities’ of the tools:
The approach to poverty based on the NBI would define situations of a structural kind, the past evolution, […] and implies solutions that in general exceed the level of the individual. Considering the criterion of the poverty line means taking into account its direct relationship with current household income. Thus, it is very sensitive to large fluctuations experienced by real wages in our country.Footnote 57
The idea of combining two methods was linked to the assumption that there were different historical strata of poverty that had to be distinguished and that such tools would reveal this fact:
The approach to the problem of poverty must be based on the recognition of its heterogeneity. Historically, a relatively small group of the Argentine urban population was not able to satisfy a series of needs identified as basic needs according to the cultural context. However, one of the most obvious signs of the worsening of the economic and social crisis experienced by our country during more than one decade has been the deterioration of the income of other groups. […] The measurement of poverty therefore requires two approaches: one based on the poverty line, and one based on the satisfaction of basic needs.Footnote 58
From then on, the idea of the ‘heterogeneity of poverty’, proposed by the IPA, became a topic for academic and expert discussions. Based on this notion, and as a consequence of exchanges with experts from ECLAC, Argentine technicians defined two new types of populations: the ‘structural poor’ and the ‘impoverished poor’, which soon became the ‘new poor’.Footnote 59 , Footnote 60 These categories joined the ‘chronic’ poor and the ‘non-poor’ (see Table I). The combination of methods allowed the definition of new social groups, which, in the context of the late 1980s and early 1990s, served to explain the crisis and the ‘decline’ of some groups of Argentine society, represented as a society that was undergoing progressive fragmentation and was becoming more unequal. According to the conclusions of the IPA:
The impact of the economic crisis changed the composition of different social groups that make up our society: the historically and structurally poor sectors, which had suffered from the beginning the current economic vicissitudes, were joined by others, setting at the same time a complexity that adds up to the normal heterogeneity of the sectors that live in a state of scarcity.Footnote 61
Table 1. The Heterogeneity of Poverty According to the IPA
Source: IPA/INDEC, Sobre la pobreza en Argentina, pp. 18–19; El método de las necesidades básicas insatisfechas (NBI) y sus aplicaciones en América Latina, p. 29.
The ‘new poor’ were, in effect, those who fell: although they were unable to reach a threshold of minimum consumption, they lived, by virtue of a more prosperous past, in ‘structurally’ non-poor households that had access to basic housing and maintained a degree of access to the health and education systems deemed normatively acceptable.Footnote 62 With this new concept, which defied the traditional tripartite classification of society according to social class, poverty became a category that reconfigured the previously dominant representation of society and its social divisions. The expert classification of the ‘new poor’ was even assimilated, especially among the middle class, as a form of self-representation.Footnote 63
However, in the 1990s, as poverty was established as a public problem, as a subject of political dispute, and as a field of expert intervention, the income method gradually replaced the combined approach: relying on information collected on a regular basis by the EPH surveys, and therefore updatable several times per year, the income-based method made it possible to track short-term variations in poverty.Footnote 64 Monitoring the social situation of the country, the results of public policies, in general, and of social policies, in particular, became much easier. Additionally, the income method became a source of debate in the news and in the media, viewed as a ‘thermometer’ of the social question, and it was integrated into academic, political and journalistic arguments. Thus it became a tool for evaluating the performance of the government in the social field. At the same time, the active promotion of the ECLAC contributed to consecrating the income method as an official measure available to national states, while the discussions on the specific methods remained limited to specialised circles.
The IPA was dissolved shortly after Carlos Menem's government took power, when his secretary of planning, Moisés Ikonikoff, made clear the official lack of interest in maintaining this expert project. Vinocur and Minujin left the institute and were hired by the regional office of UNICEF in Buenos Aires; Beccaria resigned from the direction of the INDEC. With the departure of the majority of the experts involved in the IPA, a part of the knowledge and capabilities accumulated in the course of these experiences departed together with their bearers; another group would be transferred to new state agencies created in the 1990s. Former experts from the IPA designed the SIEMPRO, ‘thinking of recreating the IPA’.Footnote 65 There, with the support of the Secretariat for Social Development and with funding from the IDB and the UNDP, they carried out the task of elaborating indicators on poverty, which became a separate, independent, expert domain.Footnote 66
Although some portion of the technical-bureaucratic capacities generated in the course of this process did not remain with the public agency responsible for producing official statistical data, the project dedicated to the elaboration of ways to measure poverty in Argentina and the corresponding classification of the social world left a strong legacy. Even when the succession of different methods of measurement was somewhat chaotic and conflictive, the construction of the ‘social fact’ of poverty remained robust. Poverty numbers became a tool for political struggle at a time when other traditional resources of political discourse, such as appealing to alternative economic models, were in retreat, and they provided an instrument to formulate critiques of the neoliberal reforms of the 1990s. Vinocur and Minujin occupied important posts at UNICEF, the first multilateral organisation that criticised the Menem government's policies of structural adjustment. During the first half of the 1990s, in a context of monetary stability and high economic growth rates which politically strengthened the government, providing statistical information on poverty and infant mortality was, according to Minujin, ‘a way of showing the dark side’ of the situation.Footnote 67 Towards the beginning of 2000, during the economic, social, and political crisis experienced by Argentina, decoding the dismal situation in terms of poverty allowed the political narratives and the media to shed light on the magnitude of the difficulties experienced by a large sector of society.Footnote 68
Conclusion: The Legacies of a Founding Enterprise
The institutionalisation of the measurement of poverty in Argentina during the democratic transition shows how statistical tools can articulate a durable consensus, if not in terms of how to deal with social problems, but at least on the valid modes of accounting for them, that is, of framing them. Synthesised in a number, poverty (as a convention-based statistical object) gained considerable weight in the public sphere. In the political debate, it was established as a ubiquitous representation of Argentine society and its problems.
The programme launched in 1984 pioneered the establishment of poverty as a measurable phenomenon, at a time when the category poverty was not yet fully agreed upon in the expert field, contributing thus to its definition as an object of specific policies. Later, the IPA project made it possible to shed light on the heterogeneity of situations of poverty and its historical layers. With respect to their impact on ordering the social world, the original instrument based on NBI allowed the framing of the social ‘legacy’ of dictatorship, while the newly elaborated measurement that articulated NBI with PL made intelligible the ‘decline’ of Argentine society in times of economic crisis, such as the one that led to hyperinflation in 1989. While hyperinflations disrupted life in society, poverty as a statistical category organised the social perception of its effects.Footnote 69
Another legacy of these pioneering undertakings can be observed at the institutional level. The creation in 1993 of the Ministry of Social Development (Secretaría de Desarrollo Social, SDS), whose main objective was to combat poverty, embodied the state's response to a problem refracted in public statistics. Poverty numbers changed the institutionalised modes of thinking about the social question. The SDS, which benefited from a series of international funds associated with the promotion of targeting and new ‘social management’, relied on the existence of a state agency (called SIEMPRO) that was producing social knowledge and was inspired by the IPA.
It would require a much more extensive study to explain why poverty had not emerged earlier as a public issue or as a relevant object of public policy in Argentina. The literature has tended to see the genesis of the quantification of poverty in Argentina through the lens of the protagonists, coinciding with the idea that it was the ‘visible’ consequences of the social deterioration of the economic policies implemented by the last military dictatorship that ‘forced’ official agencies to embark on the systematic production of information on poverty.Footnote 70 According to this view, the interest in counting poor households and determining the number of poor people would be explained, on the one hand, by a transformation of the economic structure that was the consequence of an unpopular authoritarian project and, on the other hand, by an intuition concerning the propagation of poverty in the country, a phenomenon until then considered residual or of relatively moderate proportions. The empirical evidence collected in our study shows that, in order to comprehend how poverty was installed as the core of the social question, it is necessary to focus the analysis on the activity of the experts and their ability to mobilise political considerations and resources from state and multilateral agencies, for framing, quantifying, and defining the social problems on the political agenda.
The Argentine experience, perhaps because of the sense of a radical change and the emergence of a new situation, served as a laboratory for the implementation of methods of measurement in Latin America, which made it into a model to be followed by other countries in the region. However, the role of international agencies in providing the necessary resources to carry out the projects on poverty measurement cannot be overlooked. In this sense, they contributed to uprooting the traditional ways of considering the social question and inaugurating the primacy of conceptualising it in terms of poverty. However, the case we have discussed shows that it is difficult to understand the relationship between the state and transnational agencies as a unilateral linear influence, given the pragmatic relationship that the local experts established with these sources of funding, the margins of autonomy of their discussions, and the different types of measurement that they ended up adopting.
Finally, it is worth recalling that in the 1990s, while the category of poverty and the instrument for its quantification became more and more accepted among the general public, the newly constituted expert field abounded in criticisms against the established modes of measuring poverty, questioning the validity of those tools, and even looking for alternative instruments. In fact, this search for alternatives led to the creation of the Executive Committee for the Study of Poverty in 1993. The experts on the Committee spoke out against the dispersion of the indicators of poverty and highlighted the fragility of the instruments in use until then. The regional and international debates have continued to the present day, as is evidenced by a lively competition among experts linked to different schools. Nonetheless, beyond the disputes among specialists, the phenomenon, which was still very vaguely defined in the 1980s, emerged as a statistical object of great importance, as well as a concept central to classifying and ordering Latin American societies and their problems.
In this context, one can understand the importance of the political intervention in Argentine public statistics that occurred in 2007. Pointing out the conventional character of statistical objects does not imply denying its materiality and its importance as a principle of social intelligibility. The conventions therefore operate as ways of stabilising the representations of society. To interrupt such a consensus, which involves, as we have seen, both expert communities and other political and social actors, without proposing new ways of reading the social question, implies depriving society, and the state, of the ways of thinking of itself and of the conceptual tools for turning itself into a subject of discussion in democratic public space.
